Some Thoughts on Insincere Voting


By: Assistant Professor Francis Joseph Dee

TL;DR [1]

What is an insincere vote?
An insincere vote is when a voter casts a vote for a candidate who is not their most preferred candidate.

Why cast an insincere vote?
The most common reason to cast an insincere vote is because the voter is trying to affect the outcome of the election but their most preferred candidate seems unlikely to win. In cases like this, a sincere vote is unlikely to affect who wins the election.

Is it unreasonable to vote for a candidate unlikely to win?
No. A voter might believe that their vote is unlikely to affect the result of the election no matter what. Another might believe that the leading candidates are too similar, which means their vote is unlikely to affect the result substantively. Both may choose instead to take the election as an opportunity to express their support for their most preferred candidate.

Is it unethical to cast an insincere vote?
I believe that the only unethical voter is one who doesn’t do their best to bring about the good of society when voting. The latter can involve preventing a candidate they believe to be harmful to society from winning by casting an insincere vote or expressing support for a deserving but unpopular candidate with a sincere vote.

Defining insincere voting
An election involves making a single choice for a group based on the choices of its individual members. An electoral system is a procedure that derives a group choice from individual choices. Philippine presidents are elected by plurality rule, where each individual voter casts a vote for one candidate, and the candidate who gets the most votes is chosen for the presidency.

Casting the vote for a candidate is an action, while a candidate winning is an outcome. A voter can be expected to have preferences among the different possible outcomes of the election. These preferences can be expressed as an ordering for each voter. The voter can also choose among a range of actions on election day. For example, a voter in 2016 may have had a.) Roxas winning as their most preferred outcome, followed by b.) Poe winning, c.) Santiago winning, d.) Binay winning, and e.) Duterte winning in that order (a > b > c > d > e). On election day, that voter can choose one of five actions: 1.) voting for Roxas, 2.) voting for Poe, 3.) voting for Santiago, 4.) voting for Binay, or 5.) voting for Duterte.

In the context of plurality rule, a voter casts a sincere vote when the candidate that they vote for is the same as their most preferred outcome’s winning candidate (e.g., A voter wants Roxas to win the most and votes for Roxas).

On the other hand, if a voter casts a vote for a candidate who is not the winner in their most preferred outcome, they cast an insincere vote (e.g., A voter wants Roxas to win the most but votes Poe.). There are as many reasons to cast an insincere vote, but one reason may be because voters are not indifferent among all the other possible outcomes when their most preferred outcome does not occur, and the availability of polling data and other information makes it possible to make that prediction (e.g., A voter wants Roxas to win most but prefers Poe winning to Duterte winning and votes for Poe rather than Roxas since Poe and Duterte are ahead in the polls.). When the motive of casting an insincere vote is to increase the chance of affecting the outcome of the election, then that vote is a strategic insincere vote [2].

How prevalent is strategic insincere voting?
Because a method to identify a strategic and/or insincere vote requires inferring the motives of the voter that cast it and tailoring the method depending on the electoral system, there isn’t a consensus in political science on measuring how prevalent these votes are in a given election. About 3% of voters were estimated to have cast a strategic insincere vote in the 1997 Canadian elections (Blais et al. 2001); about 33% of voters were estimated to not have purely expressive (see below) voting intentions across the 2005 and 2009 German elections (Spenkuch 2018); while 63%-85% of votes were estimated to have been strategic but only 1%-3% of votes were estimated to have been insincere in the 2005 Japanese elections (Kawai and Watanabe 2013). In short, estimates of strategic voting vary widely, but it seems consistent that only a minority of voters vote insincerely.

In the context of British, French, and German elections from 2005-15, holding a university degree was not found to influence whether one votes for the leading two candidates, while male and young voters were found to be more likely to vote for candidates other than the top two (Kroeber et al. 2020). The question of whether such votes can still be strategic will be discussed next.

Do sincere voters ignore strategy?
When sincere votes are for the eventual winner of the election or for candidates neck-in-neck with the eventual winner, such votes are trivially strategic as well. In fact, some scholars even take the extreme step of classifying votes that are not for the top two best performing candidates as “wasted votes” (e.g., Kroeber et al. 2020). I disagree with this view and argue that there is at least one other scenario where a voter can make a sincere vote after making strategic considerations.

It was stressed above that voters have preferences over electoral outcomes. Implicitly, it was also assumed that voters do not have preferences over their actions. This means that theoretically, a voter whose most preferred outcome is a Roxas victory would be indifferent between a world where they voted for Roxas and Roxas won and a world where they voted for Poe and Roxas won. However, one can imagine that there are voters for whom this is not the case.

Voters who value voting sincerely can be said to “derive expressive utility” (Spenkuch 2018: 73, emphasis mine) from their vote. From there, one can imagine a voter who only cares about expressive utility regardless of outcome and will thus always vote sincerely. For the purposes of this article, such voters are considered non-strategic. While such voters may be called naïve, others may argue that it is more naïve to believe that a single vote can alter the outcome of an election, making expressive utility the only realistic motivation to vote. I agree with neither position but concede that whether one votes sincerely or insincerely, elections require some idealism of voters.

Now, the non-trivial strategic sincere voting scenario occurs when voters are motivated both by expressive utility and strategic considerations (i.e., the chance to affect election results), which I believe applies to most of us. In this scenario, suppose that voters rank their preferences and have degrees to their preferences, which again, I believe applies to most of us. Say voter 1 prefers Santiago to Roxas by a lot and prefers Roxas to Duterte by a lot, while voter 2 prefers Santiago to Roxas by a lot but Roxas to Duterte by very little. A survey then reports that the top two candidates are Duterte and Roxas, with Santiago lagging far behind. On the day of the election, voter 1 casts an insincere vote for Roxas, but voter 2 casts a sincere vote for Santiago.

I propose that both votes are strategic given each voter’s preferences. That voter 1’s vote is strategic is uncontroversial, but I argue that for voter 2, even if they believe that their vote can change the outcome of the election, the tiny utility difference between Roxas winning and Duterte winning is outweighed by the expressive utility of voting sincerely or the cost of changing one’s decision (e.g., gathering information on other candidates, etc). While I have no empirical literature to cite discussing this type of strategic voting, similar reasoning has been applied to explain strategic abstention among ideologically extreme voters when all candidates are moderate (Lefkofridi et al. 2014).

Is strategic insincere voting ethical?
Advocates of electoral system reform may be disappointed to learn that the opportunity for strategic insincere voting is not unique to plurality systems. While some electoral systems are worse than others, the Gibbard-Satterthwaite theorem (Benoit 2000) shows that any fair and rational method for aggregating individual preferences can generate scenarios that incentivize strategic insincere voting when there are more than two candidates. Simpler electoral systems (i.e., majoritarian rather than proportional systems and plurality among majoritarian ones) may be more egalitarian as they make it easier to see when an insincere vote is strategic, but majoritarian systems’ tendency to under- represent minority groups and views should not be ignored. However, as the possibility will always exist where there are elections, societies that value electoral democracy must decide whether the practice of strategic insincere voting is acceptable at all.

No amount of evidence can settle this question, but I argue that strategic insincere voting is consistent with Aristotelian virtue ethics. Leaving his thoughts on whether elections are democratic or whether desirable (two very different questions for him) to one side, Aristotle argued that for human beings, a good life involves the use of wisdom, and that the second-best kind of life is that of the political man (Kraut 2018) because politics involves the use of practical wisdom, which is the faculty used for problems whose solutions vary from situation to situation. The framework of preference orderings described above ignores whether certain preferences are ethical, but the ethical person would use their practical wisdom to deliberate with their fellow citizens and collectively arrive at a preference ordering that best serves the community. Should deliberation fail to arrive at a consensus and a vote be required, the ethical person should then decide, again with practical wisdom, on the best way to achieve the good of the community, which may or may not involve insincere voting.

Voting wisely is easier said than done.
As if the stakes of presidential elections weren’t high enough, strategic insincere voting adds another layer of complexity to an already difficult voting decision. As such, my very tame advice to voters is to be kind to each other. Strategic insincere voters are not trapo if they’re trying to figure out the best way forward given the situation. Likewise, sincere voters are not stupid as they may be trying to use their one chance every six years to express their politics by saying “no” to other similarly depressing alternatives. One of the two may be a better way to serve the country, but finding out which one is not straightforward. Deliberate, but don’t vote-shame.

[1] I’d like to thank Vivien de Guzman, Sheila Delos Angeles, Adrian Gache, and Chelsea Estrella whose work in various classes exposed me to some insights and literature discussed here.

[2] This is an intentionally narrow understanding of strategy. Sometimes, voters cast a tactical protest vote for a less preferred party not to affect the outcome of the election per se but to punish their most preferred party for making a bad policy decision (Alvarez et al 2018). Some variants of rational choice theory understand strategy more broadly, defining strategic or rational choice as any choice that maximizes utility, regardless of whether expressive, substantive, or other considerations are considered. However, others criticize that defining rationality this broadly makes rational choice theory unfalsifiable (Whitely 1995).

Barbera, S. (2001). An introduction to strategy-proof social choice functions. Social Choice and Welfare 18: 619-53
Blais, A., Nadeau, R., Gidengil, E., & Nevitte, N. (2001). Measuring strategic voting in multiparty plurality elections. Electoral Studies 20(3): 343-52
Kawai, K. & Watanabe, Y. (2013). Inferring strategic voting. American Economic Review 103(2): 624-62
Kraut, R. (2018). Aristotle's ethics. In E. N. Zalta (Ed.) The Stanford Encyclopedia of Philosophy (Summer 2018 Edition)
Kroeber, C. Le Gall, C., & Dingler, S, C. (2020). How individuals’ social characteristics impact the likelihood to waste a vote: Evidence from Great Britain, Germany, and France. Journal of
Elections, Public Opinion, and Parties. DOI: 10.1080/17457289.2020.1718155
Lefkofridi, Z., Giger, N., & Gallego, A. (2014). Electoral participation in pursuit of policy representation: Ideological congruence and voter turnout. Journal of Elections, Public Opinion, and Parties 24(3): 291-311
Spenkuch, J. L. (2018). Expressive vs. strategic voters: An empirical assessment. Journal of Public Economics 165: 73-81
Whitely, P. (1995). Rational choice and political participation: Evaluating the debate. Political Research Quarterly 48(1): 211-233.